It isn't that I run out of things to post the to blog that slows me down. It is that I get hung up on a stupid problem at work. And I can't see taking time to post (or much else) until the stupid problem is fixed. By stupid, here, I don't mean that it is easy or hard -- just that it has nothing to do with physics. A stupid problem is one of those problems that you are stuck on because you decided to approach a really important problem slightly differently... and you thought it would be quicker and cooler in the end.
When was the last time you had to look at assembly language? The language that the CPU your computer is actually running on is using? Not the high fly, wine-drinking C++/Java/C that most people in the world write in? Last time for me was undergraduate. I had to write this @&#&^@ interrupt driven assembly language file copy program. For those of you that don't remember what this crap looks like here is a sample:
004149BF mov esi,esp
004149C1 push offset string "" (41A0C8h)
004149C6 mov eax,dword ptr [tin_one]
004149C9 mov edx,dword ptr [eax]
004149CB mov ecx,dword ptr [tin_one]
004149CE call dword ptr [edx+9Ch]
004149D4 cmp esi,esp
004149D6 call @ILT+700(__RTC_CheckEsp) (4112C1h)
It turned out that +9Ch was the key to the problem. But it was two days down the toilet. Grrr.
P.S. Details, ugly details: The problem was that the TInterpreter object was defined differently when it was built and when you built against it; so memory layout wasn't consistent.